Distribution utilities are increasingly adopting analytics products into their enterprise IT portfolios. Analytics use cases are focused towards very diverse areas including finance, enterprise and customers. Grid analytics is a piece of this puzzle that focuses specifically on utility planning and operations. These areas are mainly:
• System reliability
• Asset risk and management
• Power quality
• Distributed Energy Resource integration and management
Distribution utilities store their operational information in multiple backend systems. These include GIS for asset data, OMS for outage information, DMS/ADMS/SCADA systems for real-time data, CIS systems for customer data and with the advent of AMI, interval data for customer consumption, premise voltages and power quality. All these data sources capture historic behavior of the system. These backend systems have been traditionally managed and used within their silos for specific usage. Grid analytics is all about bringing this data to a common platform and combining them to do more advanced holistic analysis to provide much larger benefit than what is possible within the silos.
Grid analytics use cases mainly analyze enterprise-wide concerns in addition to local concerns.
Combining all the data hidden in the silos in a common analytics platform enables data democratization. Data from other groups within the utility is readily available to everyone within the utility (with the proper permissions). This allows analysis that is more holistic and crosses silo boundaries. Traditionally, utility engineers spend 50-70 percent of the time collecting data from various sources for their analyses. An analytics platform makes this data available readily to them on demand hence increasing overall productivity. Grid analytics also facilitates a large number of enterprise-wide analytics and optimization use cases using real data present within the data systems bringing in significant operational efficiency.
Typical progression of grid analytics use cases has four stages.
Visualization: Visualization of areas under stress within the grid is the first step into grid analytics. Some use cases include visualization of areas with bad reliability indexes, assets with highest risk of failure and power quality issues. Visualization is typically done using geospatial maps, tables and charts. The main goal at this stage is to identify locations within the system that need improvement. Another purpose is also to track impact of system improvements.
"Combining all the data hidden in the silos in a common analytics platform enables data democratization"
Prediction: The next stage is to predict possible outcomes based on historical performance. Prediction allows utilities to proactively solve future problems. Prediction of assets that might fail due to overloading or abnormal weather is one such use case. This allows utilities to replace assets under risk before they fail and hence improve system reliability and customer satisfaction. Prediction is done mainly using statistical or machine learning models that are trained using historical data.
Simulation: Simulation enables ‘what-if’ type of analysis by calculating or predicting system impact due to changes within system resources. Like prediction, simulation supports proactive management of the distribution system. Simulation of impact of solar PV deployments, impact of replacing at risk assets on system reliability and impact on loading due to feeder re-configuration are some of the simulation use cases.
Optimization: The eventual goal of grid analytics is to optimize the operation and planning of the entire utility system across all the silos. The optimization typically will involve financial and engineering objectives and constraints. It may target stressed areas of the system or the complete system. All the earlier stages support and lead to this large-scale optimization.
Some Use Cases
Use cases within grid analytics target specific areas focused on utility operations and planning as specified earlier in the article. Let us look at some example use cases.
Reliability Optimization: Maintaining elevated level of reliability is critical for utilities. This specific use case looks at the past outages, including those obtained from OMS and AMI system to identify worst performing areas in the utilities distribution system. The historical information is then used to determine optimized project portfolio to improve system reliability. The optimization is done either to optimize reliability metrics within a given budget or to determine optimum project combination and budget to reach specific level of reliability.
Asset Risk Analysis: Utilities own many assets including transformers, overhead and underground lines, and protection devices. Failure of these assets results in power outages, thus impacting system reliability. Number of large, critical assets (such as substation transformers) are typically instrumented and tested for their conditions. However, considerable number of assets, especially at the distribution level, are not instrumented or monitored directly. In such cases indirect calculations using meter data is used to determine the condition of such assets. Thus, Asset Risk Analysis uses direct and indirect measurements to estimate the risk of asset failure and its impact on customers. It also then determines optimal asset upgrade strategy to optimize budget or to reach specific levels of aggregate risk.
PV Hosting Capacity: With renewable generation, especially solar PV, gradually arriving at price parity with traditional generation, the amount of their deployment at distribution level is increasing. However, intermittency of these resources can cause significant issues for utilities including fluctuations in voltages and power flows. PV Hosting capacity use case determines the maximum amount of solar PV a feeder or substation can safely support without causing operational issues.
Power Quality: AMI meters and other sensors in the distribution system measure voltage and other power quality elements (e.g. voltage sag/swell, momentary outages, harmonics). These measurements are used to identify locations in the grid that experience power quality issues. Using additional data from systems such as GIS, root causes of the power quality issues such as failing distribution transformers, stuck capacitors or regulators are also identified and can thus be rectified.
In addition to the above, many other uses cases exist. There are ongoing efforts by the utility industry to define grid analytics use cases. These include efforts by utilities, Electric Power Research Institute (EPRI) and Utility Analytics Institute (UAI).
Check Out:- SemiConductor Review